• The Mathematical Education
• /
• v.56 no.1
• /
• pp.63-80
• /
• 2017

By analysing the examination papers from third grade high school students, we classified the errors occurred in the problem solving process of high school 'Geometry and Vectors' into several types. There are five main types - (A)Insufficient Content Knowledge, (B)Wrong Method, (C)Logical Invalidity, (D)Unskilled Expression and (E)Interference.. Type A and B lead to an incorrect answer, and type C and D cannot be distinguished by multiple-choice or closed answer questions. Some of these types are classified into subtypes - (B1)Incompletion, (B2)Omitted Condition, (B3)Incorrect Calculation, (C1)Non-reasoning, (C2)Insufficient Reasoning, (C3)Illogical Process, (D1)Arbitrary Symbol, (D2)Using a Character Without Explanation, (D3) Visual Dependence, (D4)Symbol Incorrectly Used, (D5)Ambiguous Expression. Based on the these types of errors, answers of each problem was analysed in detail, and proper ways to correct or prevent these errors were suggested case by case. When problems that were used in the periodical test were given again in descriptive forms, 67% of the students tried to answer, and 14% described flawlessly, despite that the percentage of correct answers were higher than 40% when given in multiple-choice form. 34% of the students who tried to answer have failed to have logical validity. 37% of the students who tried to answer didn't have enough skill to express. In lessons on curves of secondary degree, teachers should be aware of several issues. Students are easily confused between 'focus' and 'vertex', and between 'components of a vector' and 'coordinates of a point'. Students often use an undefined expression when mentioning a parallel translation. When using a character, students have to make sure to define it precisely, to prevent the students from making errors and to make them express in correct ways.

• Communications of Mathematical Education
• /
• v.32 no.3
• /
• pp.435-453
• /
• 2018

The results analysis of National Assessment of Educational Achievement(NAEA) can provide various implications for teaching and learning in the school field. In this study we analyzed the deviation of NAEA achievement level-based correct-answer rate and the distribution of the responses percentages for three years from 2015 to 2017 focused on multiple-choice items. First we calculated the deviation of correct-answer rate between advanced level and proficient level and between proficient level and basic level, and then we selected the items whose deviation of correct-answer rate is more than the third quartile. We explored what the content areas, behavior areas and achievement standards of each item are and what the achievement standards and characteristics of these questions are. Based on the results of the analysis, we intended to derive implications for appropriate teaching-learning methods at each achievement-level.

• Journal of Information Technology Services
• /
• v.20 no.4
• /
• pp.129-140
• /
• 2021

Recently, interest in AI tutors is rising as a way to bridge the educational gap in school settings. However, research confirming the effectiveness of AI tutors is lacking. The purpose of this study is to explore how effective learner characteristics and recommended item learning activities are in predicting learner's academic performance in an adaptive online learning environment. This study proposed the hypothesis that learner characteristics (prior knowledge, midterm evaluation) and recommended item learning activities (learning time, correct answer check, incorrect answer correction, satisfaction, correct answer rate) predict academic achievement. In order to verify the hypothesis, the data of 362 learners were analyzed by collecting data from the learning management system (LMS) from the perspective of learning analytics. For data analysis, regression analysis was performed using the regsubset function provided by the leaps package of the R program. The results of analyses showed that prior knowledge, midterm evaluation, correct answer confirmation, incorrect answer correction, and satisfaction had a positive effect on academic performance, but learning time had a negative effect on academic performance. On the other hand, the percentage of correct answers did not have a significant effect on academic performance. The results of this study suggest that recommended item learning activities, which mean behavioral indicators of interaction with AI tutors, are important in the learning process stage to increase academic performance in an adaptive online learning environment.

• KIPS Transactions on Software and Data Engineering
• /
• v.8 no.1
• /
• pp.27-36
• /
• 2019

In this paper, we propose a method to find the most similar answer to the user's response from the question-answer database in order to avoid generating a redundant question in retrieval-based automatic question generation system. As a question of the most similar answer to user's response may already be known to the user, the question should be removed from a set of question candidates. A similarity detector calculates a similarity between two answers by utilizing the same words, paraphrases, and sentential meanings. Paraphrases can be acquired by building a phrase table used in a statistical machine translation. A sentential meaning's similarity of two answers is calculated by an attention-based convolutional neural network. We evaluate the accuracy of the similarity detector on an evaluation set with 100 answers, and can get the 71% Mean Reciprocal Rank (MRR) score.

• Journal of Intelligence and Information Systems
• /
• v.25 no.3
• /
• pp.161-177
• /
• 2019

In this paper, we study the performance improvement of the answer extraction in Question-Answering system by using sentence dependency parsing result. The Question-Answering (QA) system consists of query analysis, which is a method of analyzing the user's query, and answer extraction, which is a method to extract appropriate answers in the document. And various studies have been conducted on two methods. In order to improve the performance of answer extraction, it is necessary to accurately reflect the grammatical information of sentences. In Korean, because word order structure is free and omission of sentence components is frequent, dependency parsing is a good way to analyze Korean syntax. Therefore, in this study, we improved the performance of the answer extraction by adding the features generated by dependency parsing analysis to the inputs of the answer extraction model (Bidirectional LSTM-CRF). The process of generating the dependency graph embedding consists of the steps of generating the dependency graph from the dependency parsing result and learning the embedding of the graph. In this study, we compared the performance of the answer extraction model when inputting basic word features generated without the dependency parsing and the performance of the model when inputting the addition of the Eojeol tag feature and dependency graph embedding feature. Since dependency parsing is performed on a basic unit of an Eojeol, which is a component of sentences separated by a space, the tag information of the Eojeol can be obtained as a result of the dependency parsing. The Eojeol tag feature means the tag information of the Eojeol. The process of generating the dependency graph embedding consists of the steps of generating the dependency graph from the dependency parsing result and learning the embedding of the graph. From the dependency parsing result, a graph is generated from the Eojeol to the node, the dependency between the Eojeol to the edge, and the Eojeol tag to the node label. In this process, an undirected graph is generated or a directed graph is generated according to whether or not the dependency relation direction is considered. To obtain the embedding of the graph, we used Graph2Vec, which is a method of finding the embedding of the graph by the subgraphs constituting a graph. We can specify the maximum path length between nodes in the process of finding subgraphs of a graph. If the maximum path length between nodes is 1, graph embedding is generated only by direct dependency between Eojeol, and graph embedding is generated including indirect dependencies as the maximum path length between nodes becomes larger. In the experiment, the maximum path length between nodes is adjusted differently from 1 to 3 depending on whether direction of dependency is considered or not, and the performance of answer extraction is measured. Experimental results show that both Eojeol tag feature and dependency graph embedding feature improve the performance of answer extraction. In particular, considering the direction of the dependency relation and extracting the dependency graph generated with the maximum path length of 1 in the subgraph extraction process in Graph2Vec as the input of the model, the highest answer extraction performance was shown. As a result of these experiments, we concluded that it is better to take into account the direction of dependence and to consider only the direct connection rather than the indirect dependence between the words. The significance of this study is as follows. First, we improved the performance of answer extraction by adding features using dependency parsing results, taking into account the characteristics of Korean, which is free of word order structure and omission of sentence components. Second, we generated feature of dependency parsing result by learning - based graph embedding method without defining the pattern of dependency between Eojeol. Future research directions are as follows. In this study, the features generated as a result of the dependency parsing are applied only to the answer extraction model in order to grasp the meaning. However, in the future, if the performance is confirmed by applying the features to various natural language processing models such as sentiment analysis or name entity recognition, the validity of the features can be verified more accurately.

• The Korean Journal of Applied Statistics
• /
• v.8 no.1
• /
• pp.9-26
• /
• 1995

In surveys on sensitive issues which respondents are unwilling to answer, response bias usually occur since respondents tend to answer untruthfully or evade answer. Warner(19650 introduced the Randomized Response Technique (RRT) which protected the privacy of the individual respondent to reduce these response biases. Though this technique are theoretically good it has some problems in applying this technique to field survey. Therefore in order to apply easily RRT to practical survey we present the practical plan through comparing and analyzing the several cases which RRT was applied. Also we take the field survey according to this plan.

• Speech Sciences
• /
• v.11 no.2
• /
• pp.151-164
• /
• 2004

This study examined the use of intonation in children with single word sentences and investigated the communicative functions of pitch range and pitch direction. Two children aged 13months were observed in interaction with their mothers for 10 months. The vocalizations were coded separately for communicative function and for prosodic feature. Results show that level tones are used most frequently, and pitch range is higher for request than declaration or answer and lower for answer than request or declaration. And trends in prosodic contours were observed in request, declaration, and answer respectively. For one child, rising tones were frequently associated with request whereas rising-falling tone with declaration. For the other child, rising-level tones were more frequently associated with request whereas falling-level appeared more often in conjunction with declaration. These trends appeared more distinct in proportion as they grow in months. This result indicate that the way to express communicative functions transfer gradually from differentiating pitch range to diversify pitch direction.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.5 no.6
• /
• pp.539-545
• /
• 2004

For question answering system that extracts an answer and output to user‘s natural language question, a process of question type classification from user’s natural language query is very important. This paper proposes a question and answer type classifier using the interrogatives and word semantic categories instead of complicated classifying rules and huge dictionaries. Synonyms and postfix information are also used for question type classification. Experiments show that the semantic categories are helpful for question type classifying without interrogatives.

• Journal of Educational Research in Mathematics
• /
• v.9 no.1
• /
• pp.111-119
• /
• 1999

This paper discusses jLogic, a mathematical logic education software developed by the author. jLogic is basically a MS-Windows based software that can construct first-order models, formulas and thet their satisfiablity. Logical formulas are easily input by a "keyboard" maintained by jLogic. A special finite model, called the "Toy World" can be visually cinstructed and modified. The user is supposed to answer the following 3 questions about the selected logical expression: 1. Is it a grammatically correct logical formula? 2. Is it a sentence that has a definite truth value? 3. Is th sentence true or false? When the user inputs his answer in the "Inspector window" and then presses the OK button, jLogic instantly tests the validity of the answer and tells the user the result. jLogic is freely downloaded from http://gauss.kyungpook.ac.kr/~jlogic/

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1995.10b
• /
• pp.279-284
• /
• 1995

In CAI, it is very important to evaluate the grade of understanding which students reach about the scope of problem which students are studying. In this paper, to find out students' learning achievement, we make students reply to test which the system presents and then lead evaluation result using fuzzy number about answer result. Besides, we define the degree of prior knowledge of studentsd as conditioned fuzzy number and use existing fuzzy accuracy production function begore the stage of using fuzzy number, Next, we apply conditioned fuzzy number to accuracy degree of answer produces by this function. Through this, we come to the conclusion that evaluation result as to the same answer result is changed according to the degree of prior knowledge about the scope which students are studying.